On this page
Educational games promise learning that feels like play. Parents wonder whether they are a good use of screen time, and students wonder whether a maths or physics game counts as studying. Several large reviews have looked at this, and the answer is a qualified yes: games can help, modestly, and design matters more than the label "game".
What did the big reviews find?
Wouters and colleagues (2013). Pooling 77 comparisons with 5,547 learners, they found serious games were more effective than conventional instruction for learning (d = 0.29) and retention (d = 0.36), but not significantly more motivating. Games did better when supplemented with other instruction, used over multiple sessions, and played in groups.
Clark, Tanner-Smith and Killingsworth (2016). For students from school to undergraduate level, digital games improved learning compared with non-game conditions (g = 0.33). Comparing enhanced versions of games with standard versions also showed gains (g = 0.34). Effects varied with game mechanics and with research quality.
Sitzmann (2011). Looking at computer-based simulation games in workplace training, trainees gained more knowledge and retained more than comparison groups, but the review found strong evidence of publication bias. Games worked better as a supplement, when they taught actively and when trainees could replay them. When the comparison group was also actively engaged, trainees learned less from the games.
Maths specifically. Tokac, Novak and Thompson pooled 24 studies of school maths games and found a small, marginally significant benefit, with results varying widely.
Why "games are more motivating" isn't the main story
Many people assume games work because students enjoy them. Wouters and colleagues did not find a significant motivation advantage. Richard Mayer's review in the Annual Review of Psychology suggests where gains may come from instead: studies that added one feature to a game point to five promising features, including personalisation, pretraining, coaching and self-explanation, and science, mathematics and second-language learning look like promising areas where games may beat conventional media. In other words, games that make you think, explain and get feedback look most useful; fun alone is not the point.
A checklist for judging a learning game
Use these questions, most of them drawn from the review findings above.
Is the content the gameplay? You should be using the physics or maths to win, not answering a quiz between unrelated levels.
Does it make you explain or predict? Self-explanation was one of Mayer's promising features.
Does it give feedback on why, not just right or wrong? Coaching was another.
Can you replay it? Sitzmann found learners did better when they could return as often as they wanted.
Does it fit with your teaching? Games worked better alongside other instruction, not as a replacement.
Is it more than one session? Wouters found multiple sessions helped.
If a game fails most of these, treat it as a break, not as study.
How to use a game in a study week
A simple pattern for a school or JEE/NEET student:
Learn the topic first from your class or book.
Play the game for 15 to 20 minutes, aiming at the levels that use today's topic.
Close it and write answers to three questions: what rule did the game use, where did you fail, and why?
Solve two textbook problems on the same idea. If you can, the game helped. If not, the problems are your next step.
For a related idea, see our piece on desirable difficulties: learning that feels a little hard tends to stick.
Where Fermi fits
The Fermi Simulation Lab is not a game, but it has interactive simulations for ideas easier to see than to read, and the same rules apply: predict, test and explain. Our article on what PhET research says simulations are good for explains how to get the most from them.
